RESEARCH ARTICLE Open Access

Activation and function of receptor tyrosinekinases in human clear cell renal cellcarcinomasQing Zhang1, Jian-He Liu2, Jing-Li Liu1, Chun-Ting Qi1, Lei Yan1, Yu Chen1 and Qiang Yu1*

Abstract

Background: The receptor tyrosine kinases (RTKs) play critical roles in the development of cancers. Clear cell renalcell carcinoma (ccRCC) accounts for 75% of the RCC. The previous studies on the RTKs in ccRCCs mainly focused ontheir gene expressions. The activation and function of the RTKs in ccRCC have not been fully investigated.

Methods: In the present study, we analyzed the phosphorylation patterns of RTKs in human ccRCC patientsamples, human ccRCC and papillary RCC cell lines, and other kidney tumor samples using human phospho-RTKarrays. We further established ccRCC patient-derived xenograft models in nude mice and assessed the effects ofRTKIs (RTK Inhibitors) on the growth of these cancer cells. Immunofluorescence staining was used to detect thelocalization of keratin, vimentin and PDGFRβ in ccRCCs.

Results: We found that the RTK phosphorylation patterns of the ccRCC samples were all very similar, but differentfrom that of the cell lines, other kidney tumor samples, as well as the adjacent normal tissues. 9 RTKs, EGFR1–3,Insulin R, PDGFRβ, VEGFR1, VEGFR2, HGFR and M-CSFR were found to be phosphorylated in the ccRCC samples. Theadjacent normal tissues, on the other hand, had predominantly only two of the 4 EGFR family members, EGFR andErbB4, phosphorylated. What’s more, the RTK phosphorylation pattern of the xenograft, however, was different fromthat of the primary tissue samples. Treatment of the xenograft nude mice with corresponding RTK inhibitorseffectively inhibited the Erk1/2 signaling pathway as well as the growth of the tumors. In addition, histologicalstaining of the cancer samples revealed that most of the PDGFRβ expressing cells were localized in the vimentin-positive periepithelial stroma.

Conclusions: Overall, we have identified a set of RTKs that are characteristically phosphorylated in ccRCCs. Thephosphorylation of RTKs in ccRCCs were determined by the growing environments. These phosphorylated/activated RTKs will guide targeting drugs development of more effective therapies in ccRCCs. The synergisticalinhibition of RTKIs combination on the ccRCC suggest a novel strategy to use a combination of RTKIs to treatccRCCs.

Keywords: Receptor tyrosine kinases (RTKs), Activation and function, Clear cell renal cell carcinomas (ccRCCs),Targeted therapy, PDGFRβ, Stroma cells

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

* Correspondence: qyu@sibs.ac.cn1Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555Zuchongzhi Road, Room 2-224, Shanghai 201203, ChinaFull list of author information is available at the end of the article

Zhang et al. BMC Cancer (2019) 19:1044 https://doi.org/10.1186/s12885-019-6159-2

BackgroundKidney cancers are common in developed countries and arenotoriously difficult to be treated. Ninety percent of kidneycancers are renal cell carcinomas (RCCs) which originatefrom tubular structures of the kidney. They are subdividedinto clear cell carcinoma (ccRCC), papillary carcinoma, chro-mophobe, and oncocytoma. The remaining 10% are transi-tional cell carcinomas, which are derived from cells liningthe renal pelvis and ureter [1, 2]. Standard treatments forRCCs are surgery (partial or total nephrectomy) for localizedkidney cancer, targeted therapies and immunotherapies formetastasized cancer. Seventy-five percent of the RCCs areccRCCs which are poorly sensitive to traditional chemother-apy. Targeted therapies are also limited by the lack of know-ledge of genetic mutations in the ccRCC cells.The receptor tyrosine kinases (RTKs) are a large family of

transmembrane receptors with 58 members in human [3].The ligand-induced dimerization of the RTKs lead to phos-phorylation/activation of the receptors as well as the down-stream signaling molecules [4, 5]. RTKs play critical roles inthe development of many diseases, especially cancer. Dysre-gulations of the RTK signaling through point mutation,gene amplification, overexpression, chromosomal alter-ations, and/or constitutive activation are key factors inoncogenesis [4, 6–11]. However, the activation and functionof the RTKs in ccRCC have not been fully investigated.Previous studies in ccRCCs have mainly focused on

RTKs gene expressions [12, 13]. No genetic mutations ofRTKs have been reported in the ccRCCs. The only mo-lecular mechanism related to RTKs in ccRCCs is dysreg-ulation of the pVHL/HIF axis [14, 15], which drivesexpression of VEGF and PDGFβ and, hence, activationof their receptors VEGFR2 and PDGFRβ [16–20]. There-fore, current treatments for ccRCCs are mostly anti-angiogenic tyrosine-kinase inhibitors (TKIs) targetingVEGFR, which include pazopanib, sunitinib, axitinib, so-rafenib, and bevacizumab [21, 22].In the present study, we analyzed the phosphorylation/acti-

vation/ patterns of RTKs in 10 ccRCC patient samples, 4RCC cell lines, and 4 other kidney tumor samples. Our datarevealed that multiple RTKs were activated in the ccRCCsand the phosphorylation patterns of the RTKs in the ccRCCpatients were similar to each other but different from adja-cent normal tissues and the other kidney tumors. Treat-ments with a combination of RTK inhibitors based on theirphosphorylation patterns in the ccRCC-derived xenograftseffectively inhibited the cancer cell growth. These data sug-gest an effective therapeutic strategy to treat ccRCC patients.

MethodsCollection of primary kidney tumorsThe renal tissue specimens were collected in compliancewith local ethics regulations at the Department of Ur-ology, Xin Hua Hospital Affiliated to Shanghai Jiao Tong

University School of Medicine, China. The 10 ccRCCpatients were five males and five females (Table 1). Themean age at diagnosis was 65 ± 9. The patient informa-tion of 3 other kidney cancer samples and 1 benign renaltumor sample are in Table 2. After surgical resection,tissue samples were lysed in lysis buffer (R&D Sytems,AYR001B) for protein lysates on the ice or fixed in neu-tral buffered formalin (10%) for histology staining, or im-mediately processed to establish ccRCC patient-derivedxenograft models in nude mice.

Cell lines786–0(CRL-1932), A-498(HTB-44), ACHN(CRL-1611), andCaki-1(HTB-46) cell lines were obtained from ATCC. 786–0and Caki-1 cell lines were derived from human primaryccRCC. A-498 and ACHN cell lines were derived from hu-man primary papillary RCCs. 786–0 and ACHN cells werecultured in RPMI 1640 Medium (Gibco) with 10% FBS(Gibco). A498 cells were cultured in Dulbecco’s Modificationof Eagle’s Medium (Gibco) with 10% FBS. Caki-1 cells werecultured in McCoy’s 5A Medium (Sigma) with 10% FBS.

HE stainingFixed tissues were dehydrated using grades of ethanol(70, 80, 90, 95, and 100%). Dehydration was followed byclearing the samples in two changes of xylene. The sam-ples were then impregnated with two changes of moltenparaffin wax, embedded, and blocked out. The tissuesections (7 μm) were stained with hematoxylin-eosin bystandard procedures. Stained sections were observed andphotographs were taken using a Leica microscope.

RTK phosphorylation/activation profilingHuman phospho-RTK arrays (R&D Systems,AYR001B) were used according to the manufac-turer’s instructions. Briefly, a total of 5 mg proteinlysates of in vitro cultured cells, or 10 mg proteinlysates of clinical samples and mouse xenograftswere diluted in the kit-specific dilution buffer and

Table 1 Patient information of renal cell carcinoma (RCC)

No. Age Histopathology Stage

RE0370 72 Clear cell RCC II

RE0380 56 Clear cell RCC I~II

RE0390 73 Clear cell RCC II

RE0400 77 Clear cell RCC II

RE0410 67 Clear cell RCC II~III

RE0440 66 Clear cell RCC II

RE0450 53 Clear cell RCC I

RE0480 54 Clear cell RCC II

RE0490 56 Clear cell RCC II

RE0510 77 Clear cell RCC II

Zhang et al. BMC Cancer (2019) 19:1044 Page 2 of 13

incubated with blocked membranes overnight. Themembranes were washed and incubated with anti-phospho-tyrosine-HRP antibody for 2 h. The mem-branes were washed and exposed to chemilumines-cent reagent. The arrays were photographed usingImage Station 4000MM PRO system (Carestream). Thepixel densities of various spots were collected and quanti-fied with its software. The average signal (pixel density) ofthe pair of duplicate spots was determined for each RTK.A signal from the PBS negative control spots was used asa background value. And signals of reference spots in thecorners were used for normalization among different ar-rays. Phospho-RTK relative value was calculated accordingto the following formula: Phospho-RTKx relative value = (INTx-INTnc)/(INTref-INTnc). INTx is the pixel densityof RTKx, INTnc is the pixel density of background,andINTref is the density of reference spots.

Western blottingProteins were separated by SDS-PAGE and transferred to anitrocellulose membrane. The membrane was blocked in TBScontaining 0.1% Tween 20 (TBST) and 5% nonfat milk for 1 hat room temperature and then incubated overnight in TBSTcontaining 5% bovine serum albumin and primary antibodies.Membranes were then washed with TBST and incubated withhorseradish peroxidase-conjugated secondary antibody for 1 h,and immune complexes were detected by immobilonWesternchemiluminescent HRP substrate (WBKLS0500, Millipore).Primary antibodies are anti-phospho-EGFR (#3777), anti-EGFR (#4267), anti-phospho-PDGFRβ (#3161), anti-PDGFRβ(#3169), anti-phospho-InsulinRβ (#3024), anti-InsulinRβ(#3025), anti-phospho-VEGFR2 (#2474), anti-VEGFR2(#9698), anti-phospho-Met (#3077), anti-Met (#3148), anti-phospho-Akt (#4060), anti-phospho-Erk1/2 (#4370). All anti-bodies were purchased from Cell Signaling Technology. Themembranes were photographed using Azure Biosystems(c300) and were quantified using its software (Analysis Tool-box). The density ratio of interest proteins to GAPDH or β-Actin were calculated.

Xenograft models and treatmentThe female BALB/c nude (nu/nu) mice were purchasedfrom Beijing Vital River Laboratory Animal Technology Co.,Ltd. and used for implantation at the age of 6–8weeks. They

Table 2 Patient information of the other kidney cancers and abenign renal tumor

No. Age Histopathology

RE0020 59 Papillary RCC

RE0150 55 Oncocytoma

RE0210 52 Renal pelvic carcinoma

RE0500 52 Cystic nephroma

Fig. 1 A gross presentation and HE staining of a representative ccRCC total nephrectomy sample and its adjacent tissue. a. A typical grosspresentation of ccRCC with a bright yellow color. b. The adjacent normal tissue. c. HE staining of a section of the ccRCC with transparent emptycytoplasm and well-defined cell borders. d. HE staining of a section of the adjacent tissue with normal glomerulus, proximal convoluted tubules,and distal convoluted tubules. Scale bars represent 100 μm

Zhang et al. BMC Cancer (2019) 19:1044 Page 3 of 13

were maintained under specific pathogen-free conditions,and food and water were supplied ad libitum. Housing andall procedures were performed according to protocols ap-proved by the Ethics Committee of Shanghai institute ofmateria medica. Subcutaneous xenografts were establishedby injection of 5× 106 cells or one piece (2mm3) tumor permouse to right flank. Tumor formation was monitored eachweek. Each subcutaneous tumor was measured using a cali-per, and tumor volumes were calculated as follows: 0.5×length× width2. Nude mice with ccRCC patient-derived xe-nografts of approximately 100mm3 were allocated randomlyinto 4 experimental groups and orally treated with 3mg/kg/d Crizotinib (n= 6), 30mg/kg/d Lapatinib (n = 6), combin-ation of Crizotinib and Lapatinib(n = 6), or vehicle (n = 6) for21 days. Mice were euthanized in a CO2 chamber for 2 h

after the last treatment. Crizotinib and Lapatinib were pur-chased from Selleck Chemicals.

Immunofluorescence stainingCryosections were blocked in PBS containing 5% normaldonkey serum for 2 h at room temperature. Sectionswere incubated over night at 4 °C with the primary anti-bodies against PDGFRβ (ab32570, rabbit Anti-PDGF Re-ceptor beta antibody, 1:50, Abcam), Pan-Keratin (#4545,mouse anti-pan-keratin antibody,1:50, CST), Vimentin(sc-7557, goat anti-vimentin antibody, 1:50, Santa Cruz).After washed with PBS three times, the sections were in-cubated for 1 h at room temperature with Alexa Fluor594-labeled donkey anti-rabbit IgG (A21207,1:400, Invi-trogen), Alexa Fluor 488-labeled donkey anti-mouse IgG

Fig. 2 Patterns of phospho-RTK in 10 pairs of human ccRCCs and adjacent tissues. Each RTK was in duplicate. Positive control spots are locatedon the top left, top right, and bottom left of each array. (1. EGFR; 2. ErbB2; 3. ErbB3; 4. Insulin R; 5. HGFR (Met); 6. PDGFRβ; 7. M-CSFR; 8. VEGFR1; 9.VEGFR2; 10. ErbB4)

Zhang et al. BMC Cancer (2019) 19:1044 Page 4 of 13

(A21202,1:400, Invitrogen) and Alexa Fluor 555-labeledrabbit anti-goat IgG (A21431,1:400, Invitrogen). Sectionswere washed three times in PBS, followed by mountingtissue with Dako fluorescence mounting medium. Pho-tographs were taken using a Leica DMi8.

Statistical analysisData were represented as mean ± SEM. T test was usedin human phospho-RTK studies. Two-way ANOVA withTukey post hoc test was used in mouse xenograft treat-ment studies. Statistical significance was established forP < 0.05, P < 0.01, and P < 0.001.

ResultsPathological examination of the ccRCCs and theiradjacent tissuesTo examine the histopathology of the kidney tu-mors, HE staining was performed. Gross examin-ation of the resected tumor samples revealed thatthe ccRCCs were all bright yellow in color, due totheir intracellular lipid accumulation (Fig. 1a). Incontrast, the adjacent normal tissues of the ccRCCsshowed normal flesh color (Fig. 1b). In HE stainingsections, the ccRCC cells showed transparent andempty (water clear) cytoplasm with well-defined cell

Fig. 3 The relative levels of the phospho-RTKs in human ccRCCs and adjacent tissues. The phospho-RTK levels were measured using the humanphospho-RTK array kit. P < 0.05 (*), P < 0.01 (**), and P < 0.001(***) vs. adjacent tissues of clear cell RCC. Data were represented as mean ± SEM

Zhang et al. BMC Cancer (2019) 19:1044 Page 5 of 13

borders (Fig. 1c). The nuclei of ccRCCs were round.Architecturally, the ccRCCs displayed compact-alveolar or acinar growth patterns. The small nestswere surrounded by a well-developed network ofthin-walled vessels. An abundance of extravasatedred blood cells were observed in the tumors. Theglomerulus, proximal convoluted tubules, and distalconvoluted tubules in the cortex of the kidney couldbe seen in adjacent tissues (Fig. 1d).

The phosphorylation patterns of the RTKs in the ccRCCpatient-derived tumors were similarTo understand the expression and phosphorylation of theRTKs in the ccRCCs, we analyzed 10 pairs of primary ccRCCs

and their adjacent non-tumor kidney tissues using humanphospho-RTK arrays which evaluate the relative phosphoryl-ation levels of 49 receptor tyrosine kinases (Additional file 1:Fig. S1). 9 RTKs (EGFR1–3, Insulin R, PDGFRβ, VEGFR1,VEGFR2, HGFR, and M-CSFR) were found to be phosphory-lated in the ccRCC samples (Fig. 2 and Fig. 3). Comparing totheir adjacent normal tissues, Insulin R, HGFR, PDGFRβ, M-CSFR, VEGFR1, and VEGFR2 were specific to the ccRCCs.Among them, the phosphorylation levels of Insulin R,PDGFRβ, VEGFR1, and VEGFR2 were significantly increasedin all the ccRCC samples. The phosphorylation levels ofHGFR (spot #5) and M-CSFR (spot #7) varied among thesamples. HGFR was highly phosphorylated in RE0370 andRE0410 samples while M-CSFR was highly phosphorylated in

Fig. 4 Western blotting analyses of the tissue lysates of the human ccRCCs (Ca) and adjacent tissues (Ad). Tissues were lysed and protein wasanalyzed by Western blotting using antibodies as indicated. GAPDH and β-Actin antibodies were used as loading controls

Zhang et al. BMC Cancer (2019) 19:1044 Page 6 of 13

RE0370, RE0440, and RE0450 samples. This RTKs activationpatterns of ccRCCs were different from that of their paired ad-jacent tissues in which only the EGFR family members, par-ticularly EGFR and ErbB4, were significantly phosphorylated.These findings were further verified by Western blotting ana-lyses. The phosphorylation levels of Insulin Rβ (Tyr1150/1151), PDGFRβ (Tyr751), VEGFR2 (Tyr996), and HGFR (MetTyr1234/1235) were found to be increased in the tumor tis-sues in comparison to the paired adjacent tissues (Fig. 4). Inaddition, the protein levels of some of the RTKs (Insulin Rβ,PDGFRβ, VEGFR2, or Met) were also increased in certain tu-mors. The protein expression patterns of PDGFRβ andVEGFR2 in tumors were also different from their adjacent tis-sues (Fig. 4a, d).

The RTK phosphorylation patterns of ccRCC patient-derivedtumors were different from that of human ccRCC cell lines,papillary RCC cell lines, and other kidney tumor samplesTo determine whether the RTK phosphorylation pat-terns in the ccRCCs are specific, we evaluated theRTK phosphorylation patterns in 2 ccRCC cell lines,2 papillary RCC cell lines and 4 other types of kidneytumor samples. The RTK phosphorylation patterns ofthe four human RCC cell lines were similar with each

other (Fig. 5). The EGFR family and HGFR werehighly phosphorylated in all the four cell lines. Incontrast, the RTK phosphorylation patterns of thefour other types of tumor samples, namely a papillaryRCC (RE0020), an oncocytoma (RE0150), a renal pel-vic carcinoma (RE0210), and a cystic nephroma(RE0500), were different from each other and werealso different from that of the ccRCCs, except EGFR,which was highly phosphorylated in all samples(Fig.6). ErbB4, Insulin R, and IGF-1R were phosphory-lated in the papillary RCC (RE0020), Mer (Axl family)was phosphorylated in the oncocytoma (RE0150), andHGFR, PDGFRα, and PDGFRβ were phosphorylatedin the renal pelvic carcinoma (RE0210, Fig.6). In thebenign renal tumor, namely the cystic nephroma sam-ple (RE0500), only EGFR was phosphorylated (Fig.6).These data demonstrated that the RTK phosphoryl-ation patterns of the ccRCCs were specific.

The RTK phosphorylation pattern of the ccRCC sample inthe xenograft was different from that of the primarysamplesIn order to treat the tumors with tyrosine kinase inhibi-tors based on their RTK phosphorylation patterns, we

Fig. 5 Patterns of the phospho-RTKs in the human ccRCC (a) and papillary RCC (b) cell lines. EGFR (1) and HGFR (2) were all activated in the fourRCC cell lines

Zhang et al. BMC Cancer (2019) 19:1044 Page 7 of 13

tried to establish tumor xenograft models using thepatient-derived tumor samples as well as the cancer celllines. Thirty-five tissue pieces from the 10 samples ofthe ccRCCs were subcutaneously implanted into 35nude mice. Only one xenograft (RE0410) grew success-fully. We then analyzed the RTK phosphorylation pat-tern of this ccRCC explant. The RTK phosphorylationpattern of the xenograft was different from its originalprimary sample (RE0410). Only the phosphorylation ofEGFR family (EGFR, ErbB2 and ErbB3) and HGFR weremaintained at high levels while that of the other RTKsdecreased (Fig.7a). In contrast to the poor tumorigen-icity of the ccRCC samples from patients, the estab-lished cell lines of ccRCC and papillary RCC werehighly tumorigenic. Both EGFR and HGFR remainedphosphorylated in all four of the cell line-derivedxenograft samples, although their phosphorylationlevels decreased in vivo (Fig.7b, c). These data

demonstrated that the RTK phosphorylation patternsin the xenografts changed and the success rate ofsubcutaneous grafting of ccRCC samples was low innude mice.

Combination of TKIs synergistically inhibited the growthof ccRCCs in vivoPhospho-RTK array of the ccRCC explants from thexenograft mice showed that three of the EGFR familymembers and the HGFR were highly phosphorylated inthe xenograft tumors. We therefore used the RTK inhibitorstargeting EGFR family and HGFR to treat the ccRCC xeno-graft nude mice. As shown in Fig. 8a, the change of bodyweight in treatment groups was similar with that in vehiclegroup. The EGFR inhibitor lapatinib or the HGFR inhibitorcrizotinib alone slightly inhibited the tumor growth (Fig.8b).In comparison, the combination of the two inhibitors wasmuch more efficient than the single treatment to inhibit the

Fig. 6 Patterns of phospho-RTKs in the other kidney cancer samples and the benign renal tumor. The relative levels of the phospho-RTKs werecalculated and presented under each array blot

Zhang et al. BMC Cancer (2019) 19:1044 Page 8 of 13

tumor growth (Fig. 8b). The average inhibition rate of crizo-tinib, lapatinib, or a combination of them on the ccRCCwere 38.24 ± 22.40%, 35.43 ± 37.15%, and 62.79 ± 21.95% re-spectively (Fig. 8c, d).To understand the effects of the combination treatment at

the molecular level, we examined the effects of crizotinib, lapa-tinib, or the combination of them on the phosphorylation/ac-tivation of their target proteins and their downstreamsignaling molecules Erk1/2 and Akt. As shown in Fig. 8e andf, the combination treatment synergistically inhibited thephosphorylation of Met, EGFR, and Erk1/2. These data sug-gested that a combination treatment of the RTK inhibitorsbased on the RTK phosphorylation patterns synergisticallyinhibited the RTK-mediated signaling and the tumor growth.

PDGFRβ was expressed in the periepithelial stroma cellsPDGFRs are usually expressed in stroma cells. To under-stand the function of the PDGFRβ in the ccRCCs, weanalyzed the expression of PDGFRβ in the patient-derived ccRCCs and their adjacent tissues. The PDGFRβwas mainly expressed in glomerulus in the tumor adja-cent tissues (Fig. 9a). In the ccRCC tumor tissues,PDGFRβ was present in the vimentin-positive stromacells surrounding the tumor islands and blood vessels(Fig. 9b, c). But the keratin-positive epithelial cells weremainly localized in the tumor islands which werePDGFRβ-negative (Fig. 9b, c). These results suggest thatthe PDGFRβ expressing cells were periepithelial stromacells in the ccRCCs.

Fig. 7 Patterns and quantitation of the phospho-RTKs in the xenograft mice of 1 patient-derived ccRCC sample (RE0410, a), 2 human ccRCC (b)and 2 papillary RCC (c) cell lines

Zhang et al. BMC Cancer (2019) 19:1044 Page 9 of 13

DiscussionWe identified 9 RTKs that were significantly phosphory-lated in the primary ccRCC samples and 6 of which, Insu-lin R, HGFR, PDGFRβ, M-CSFR, VEGFR1, and VEGFR2,were specific to the ccRCCs samples comparing to theiradjacent normal tissues. More importantly, the phosphor-ylation patterns of the RTKs in the ccRCC patient sampleswere similar among each other. It is therefore possiblethat the activation of the 6 ccRCCs-specific RTKs are im-portant for the formation and growth of the ccRCCs. Our

data are consistent with previous studies on the expres-sions and roles of RTKs in ccRCCs. There were several re-ports demonstrated VEGF/VEGFR activation and HGFRupregulation in patients with ccRCCs [12, 17–20, 23, 24].The M-CSFR activation we observed in the ccRCC sam-ples may be due to increases and activations of the tumor-associated macrophages in ccRCCs [25–27]. The role ofInsulin R in ccRCCs is unclear [28]. There was a reportshowing that the expressions of Insulin R were similar inccRCCs and their adjacent normal tissues, but the

Fig. 8 Combination of TKIs synergistically inhibited human ccRCC growth in vivo. a and b. The body weights and tumor volumes during thedrug treatment. The ccRCC xenograft nude mice were treated with vehicle, crizotinib (Cri), lapatinib (Lap), or combination of them for 21 days.Tumors were excised and photographed at the end of treatments. c. The tumor weights at the end of treatment. D. Tumors from ccRCCxenograft nude mice. e. Western blotting analyses of P-Met, P-EGFR, P-Erk1/2 and P-Akt levels of the tumors. The numbers underneath the groupsrepresent the serial number of mice. Tumor lysates were processed for Western blot analyses and probed with the indicated antibodies. f. Theratios of protein phosphorylation levels relative to GAPDH. P < 0.05 (*), P < 0.01 (**), and P < 0.001(***) vs. vehicle group. Drug combination groupwas compared with the crizotinib group or lapatinib group (P < 0.05, #). Data were represented as mean ± SEM

Zhang et al. BMC Cancer (2019) 19:1044 Page 10 of 13

phosphorylation of the Insulin R was not analyzed in thisreport [29]. Our data demonstrated that the Insulin R wassignificantly phosphorylated in the ccRCC samples, butnot in the adjacent normal tissues, suggesting that InsulinR may have a role in promoting ccRCC cell growth. How-ever, it was also reported that Insulin R expression correlatedwith a lower Fuhrman nuclear grade and better patient prog-nosis [29]. Further studies are needed to clarify the roles ofInsulin R in ccRCCs. None the less, these data suggest thatthe 6 specifically activated RTKs in the ccRCCs may be im-portant targets for the treatment of ccRCCs.Among the 6 specifically activated RTKs, HGFR and

Insulin R were reported to be mainly expressed in theccRCC epithelial cells [23, 24, 29]. The M-CSF R seemsto be expressed in the tumor-associated macrophages[25–27] while the VEGFRs were likely expressed in theblood vessel endothelial cells. The PDGFRβ was foundto be mainly expressed within the periepithelial stromain the ccRCC samples in our study. Similar expressionpatterns of PDGFRβ were found in breast, prostate, pan-creatic, gastric, and oral squamous cell carcinoma cancercells [30–32]. More importantly, high PDGFRβ expres-sion in fibroblast-rich stroma is commonly associatedwith poor prognosis [33, 34]. These data suggest that theRTKs in the ccRCC stroma cells may also be abnormally

activated to support the growth of the cancer cells. Thus,targeting the activated RTKs in both the cancer epithe-lial cells and the surrounding stroma cells that associ-ated with poor prognosis may be a primary choice fortreating the ccRCC patients.It is unclear what caused the activation of the RTKs in

the ccRCCs. The behavior of the ccRCCs in the xeno-graft mice, however, indicated that majority of the 9RTKs might be activated by their corresponding growthfactors in the tumor environments. When the cancercells were implanted into a new environment in thexenograft mice, most of the cancer cells failed to grow,likely because of lack of necessary growth factors to acti-vate the RTKs. The only ccRCC sample that did grow inthe xenograft mouse had different RTK phosphorylationpatterns from that of the original sample. In addition,the four cancer cell lines, when implanted into the xeno-graft mice, also showed similar RTK phosphorylationpatterns as the primary cancer sample, but differentfrom that of the in vitro growing cells. All these datasuggest that the RTK phosphorylation patterns of thecancer cells are not cell autonomous, but rather are de-termined by their growing environments.Although we could not reproduce the same RTK

phosphorylation patterns of the ccRCC primary cancer

Fig. 9 Immunostaining for PDGFRβ (red), Vimentin (red) and Keratin (green) in a pair of human ccRCC tissues. Cell nucleus was stained blue by DAPI.a. Human ccRCC adjacent tissue (scale bars = 50 μm). b. Human ccRCC tissue (scale bars = 50 μm). c. Human ccRCC tissue (scale bars = 25 μm). Arrowsindicate PDGFRβ positive cells surrounding the tumor islands (*) in the ccRCC tissue. # indicates glomerulus and + indicates blood vessel

Zhang et al. BMC Cancer (2019) 19:1044 Page 11 of 13

samples in the xenograft models, the treatment of thetumor cells in the xenograft mice with a combination ofthe RTKIs, based on the RTK phosphorylation patterns,successfully inhibited the tumor cell growth, suggestingthat the RTK phosphorylation pattern-guided treatmentof cancers is an effective therapeutic strategy.

ConclusionsIn summary, we have identified a set of RTKs that are char-acteristically phosphorylated in ccRCCs. The phosphoryl-ation of the RTKs and the growth of the ccRCCs weredetermined by the growing environments of the ccRCCs.Treatment of the ccRCC xenograft mouse with a combin-ation of RTKIs based on the RTK phosphorylation patternof the ccRCC in the new environment synergistically inhib-ited the growth of the ccRCC. These data suggest a novelstrategy to use a combination of RTKIs to treat ccRCCs.

Additional file

Additional file 1: Figure S1. Schematic illustration of the RTK arrayfrom the R&D Systems. (TIF 2291 kb)

AbbreviationsccRCCs: Clear cell renal cell carcinomas; EGFR: Epidermal growth factorreceptor; HGFR: Hepatocyte growth factor receptor; IGF-1R: Insulin-likegrowth factor 1 receptor; M-CSFR: Macrophage colony-stimulating factor re-ceptor; PDGFR: Platelet-derived growth factor receptor; RTKIs: Receptortyrosine kinase inhibitors; RTKs: Receptor tyrosine kinases; VEGFR: Vascularendothelial growth factor receptor

AcknowledgementsNot applicable.

Authors’ contributionsQZ and QY contributed to conception and design of all the experiments, analysesand interpretations of the data, writing and revision of the manuscript; J-HL providedall samples and clinical data of the patients; J-LL and QZ performed the Phospho-RTK analyses; C-TQ, LY, YC and QZ performed animal experiments, HE staining, andWestern Blotting; QZ performed the immunofluorescence experiments. All authorshave read and approved the manuscript.

FundingThis work was supported in part by the China National Key Research andDevelopment Program (2018YFC1705505) and the National Natural ScienceFoundation of China (no.81673465) in the design of the study and theanalysis of data.

Availability of data and materialsThe datasets used and/or analyzed during the current study are availablefrom the corresponding author on reasonable request.

Ethics approval and consent to participateThe clinical samples were obtained from patients undergoing surgicalresection at the Department of Urology, Xin Hua Hospital Affiliated toShanghai Jiao Tong University School of Medicine, China. All theexperiments were approved by the Ethics Committee of Xin Hua HospitalAffiliated to Shanghai Jiao Tong University School of Medicine, China.Written informed consents were obtained from the patients. Samples wereconfirmed to be tumor or normal based on pathological assessment.All the protocols of the animal experiments were approved by the EthicsCommittee of Shanghai Institute of Materia Medica, and the researchcomplied with the Guide for the Care and Use of Laboratory Animals.

Consent for publicationNot applicable.

Competing interestsThe authors declare that they have no competing interests.

Author details1Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555Zuchongzhi Road, Room 2-224, Shanghai 201203, China. 2The Department ofUrology, Xin Hua Hospital Affiliated to Shanghai Jiao Tong University Schoolof Medicine, 1665 Kongjiang Road, Shanghai, China.

Received: 14 January 2019 Accepted: 13 September 2019

References1. Cohen HT, McGovern FJ. Renal-cell carcinoma. N Engl J Med. 2005;353(23):

2477–90.2. Shaw G. The silent disease. Nature. 2016;537(7620):S98–9.3. Robinson DR, Wu YM, Lin SF. The protein tyrosine kinase family of the

human genome. Oncogene. 2000;19(49):5548–57.4. Lemmon MA, Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell.

2010;141(7):1117–34.5. Maruyama IN. Mechanisms of activation of receptor tyrosine kinases:

monomers or dimers. Cells. 2014;3(2):304–30.6. Haglund K, Rusten TE, Stenmark H. Aberrant receptor signaling and

trafficking as mechanisms in oncogenesis. Crit Rev Oncog. 2007;13(1):39–74.7. Gross S, Rahal R, Stransky N, Lengauer C, Hoeflich KP. Targeting cancer with

kinase inhibitors. J Clin Invest. 2015;125(5):1780–9.8. Motallebnezhad M, Aghebati-Maleki L, Jadidi-Niaragh F, Nickho H, Samadi-

Kafil H, Shamsasenjan K, Yousefi M. The insulin-like growth factor-I receptor(IGF-IR) in breast cancer: biology and treatment strategies. Tumour Biol.2016;37(9):11711–21.

9. Zhang XY, Zhang PY. Receptor tyrosine kinases in carcinogenesis. OncolLett. 2016;12(5):3679–82.

10. Salgia R. MET in lung cancer: biomarker selection based on scientificrationale. Mol Cancer Ther. 2017;16(4):555–65.

11. Bradley CA, Salto-Tellez M, Laurent-Puig P, Bardelli A, Rolfo C, Tabernero J,Khawaja HA, Lawler M, Johnston PG, Van Schaeybroeck S. Targeting c-METin gastrointestinal tumours: rationale, opportunities and challenges. Nat RevClin Oncol. 2017.

12. Behbahani TE, Thierse C, Baumann C, Holl D, Bastian PJ, von Ruecker A,Müller SC, Ellinger J, Hauser S. Tyrosine kinase expressiong profile in clearcell renal cell carcinoma. World J Urol. 2012;30(4):559–65.

13. Thomasson M, Hedman H, Ljungberg B, Henriksson R. Gene expressionpattern of the epidermal growth factor receptor family and LRIG1 in renalcell carcinoma. BMC Res Notes. 2012;5:216.

14. Cancer Genome Atlas Research Network. Comprehensive molecularcharacterization of clear cell renal cell carcinoma. Nature. 2013;499(7456):43–9.

15. Von Roemeling CA, Marlow LA, Radisky DC. Functional genomics identifiesnovel genes essential for clear cell renal cell carcinoma tumor cellproliferation and migration. Oncotarget. 2014;5(14):5320–34.

16. Sourbier C, Srivastava G, Ghosh MC, Ghosh S, Yang Y, Gupta G, Degraff W,Krishna MC, Mitchell JB, Rouault TA, Linehan WM. Targeting HIF2αtranslation with Tempol in VHL-deficient clear cell renal cell carcinoma.Oncotarget. 2012;3(11):1472–82.

17. Shen C, Kaelin WG Jr. The VHL/HIF axis in clear cell renal carcinoma. SeminCancer Biol. 2013;23(1):18–25.

18. Smaldone MC, Maranchie JK. Clinical implications of hypoxia induciblefactor in renal cell carcinoma. Urol Oncol. 2009;27(3):238–45.

19. Baldewijns MM, van Vlodrop IJ, Vermeulen PB, Soetekouw PM, van EngelandM, de Bruïne AP. VHL and HIF signalling in renal cell carcinogenesis. JPathol. 2010;221(2):125–38.

20. Raval RR, Lau KW, Tran MG, Sowter HM, Mandriota SJ, Li JL, Pugh CW,Maxwell PH, Harris AL, Ratcliffe PJ. Contrasting properties of hypoxia-inducible factor 1(HIF-1) and HIF-2 in von Hippel-Lindau-associated renalcell carcinoma. Mol Cell Biol. 2005;25(13):5675–86.

21. Porta C, Giglione P, Paglino C. Targeted therapy for renal cell carcinoma:focus on 2nd and 3rd line. Expert Opin Pharmacother. 2016;17(5):643–55.

22. Zhi WI, Kim JJ. An update on current management of advanced renal cellcancer, biomarkers, and future directions. Ann Cancer Res. 2014;1(2):1–10.

Zhang et al. BMC Cancer (2019) 19:1044 Page 12 of 13

23. Macher-Goeppinger S, Keith M, Endris V, Penzel R, Tagscherer KE, Pahernik S,Hohenfellner M, Gardner H, Grullich C, Schirmacher P, Roth W. Met expressionand copy number status in clear-cell renal cell carcinoma: prognostic valueand potential predictive marker. Oncotarget. 2017;8(1):1046–57.

24. Gibney GT, Aziz SA, Camp RL, Conrad P, Schwartz BE, Chen CR, Kelly WK,Kluger HM. c-Met is a prognostic marker and potential therapeutic target inclear cell renal cell carcinoma. Ann Oncol. 2013;24(2):343–9.

25. Bonelli S, Geeraerts X, Bolli E, Keirsse J, Kiss M, Pombo Antunes AR, VanDamme H, De Vlaminck K, Movahedi K, Laoui D, Raes G, Van GinderachterJA. Beyond the M-CSF receptor - novel therapeutic targets in tumor-associated macrophages. FEBS J. 2018;285(4):777–87.

26. Komohara Y, Hasita H, Ohnishi K, Fujiwara Y, Suzu S, Eto M, Takeya M.Macrophage infiltration and its prognostic relevance in clear cell renal cellcarcinoma. Cancer Sci. 2011;102(7):1424–31.

27. Van Overmeire E, Stijlemans B, Heymann F, Keirsse J, Morias Y, Elkrim Y, BrysL, Abels C, Lahmar Q, Ergen C, Vereecke L, Tacke F, De Baetselier P, VanGinderachter JA, Laoui D. M-CSF and GM-CSF receptor signalingdifferentially regulate monocyte maturation and macrophage polarization inthe tumor microenvironment. Cancer Res. 2016;76(1):35–42.

28. Solarek W, Czarnecka AM, Escudier B, Bielecka ZF, Lian F, Szczylik C. Insulinand IGFs in renal cancer risk and progression. Endocr Relat Cancer. 2015;22(5):R253–64.

29. Lkhagvadorj S, Oh SS, Lee MR, Jung JH, Chung HC, Cha SK, Eom M. Insulinreceptor expression in clear cell renal cell carcinoma and its relation toprognosis. Yonsei Med J. 2014;55(4):861–70.

30. Paulsson J, Ehnman M, Östman A. PDGF receptors in tumor biology:prognostic and predictive potential. Future Oncol. 2014;10(9):1695–708.

31. Paulsson J, Sjöblom T, Micke P, Pontén F, Landberg G, Heldin CH, Bergh J,Brennan DJ, Jirström K, Ostman A. Prognostic significance of stromalplatelet-derived growth factor beta-receptor expressionin human breastcancer. Am J Pathol. 2009;175(1):334–41.

32. Kartha VK, Stawski L, Han R, Haines P, Gallagher G, Noonan V, KukuruzinskaM, Monti S, Trojanowska M. PDGFRβ is a novel marker of stromal activationin oral squamous cell carcinomas. PLoS One. 2016;11(4):e0154645.

33. Jain RK. Molecular regulation of vessel maturation. Nat Med. 2003;9:685–93.34. Floege J, Eitner F, Alpers CE. A new look at platelet-derived growth factor in

renal disease. J Am Soc Nephrol. 2008;19(1):12–23.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Zhang et al. BMC Cancer (2019) 19:1044 Page 13 of 13